GraphBridge.GraphRag.Pdf
0.2.0-preview.1
Prefix Reserved
dotnet add package GraphBridge.GraphRag.Pdf --version 0.2.0-preview.1
NuGet\Install-Package GraphBridge.GraphRag.Pdf -Version 0.2.0-preview.1
<PackageReference Include="GraphBridge.GraphRag.Pdf" Version="0.2.0-preview.1" />
<PackageVersion Include="GraphBridge.GraphRag.Pdf" Version="0.2.0-preview.1" />
<PackageReference Include="GraphBridge.GraphRag.Pdf" />
paket add GraphBridge.GraphRag.Pdf --version 0.2.0-preview.1
#r "nuget: GraphBridge.GraphRag.Pdf, 0.2.0-preview.1"
#:package GraphBridge.GraphRag.Pdf@0.2.0-preview.1
#addin nuget:?package=GraphBridge.GraphRag.Pdf&version=0.2.0-preview.1&prerelease
#tool nuget:?package=GraphBridge.GraphRag.Pdf&version=0.2.0-preview.1&prerelease
GraphBridge.GraphRag.Pdf
Experimental. A PDF loader for the GraphBridge GraphRAG toolkit, built on PdfPig (Apache-2.0). The API may change in any preview; using it reports GBRAG001.
services.AddGraphRag<AppDbContext>(embeddingSet)
.AddGraphRagPdf(); // .pdf files now load: GraphRagSources.Folder("reports", "*.pdf")
- What is stored: the PDF's text layer, page by page in reading order, one paragraph per text block. The document's bytes are that text, so citations are located in it, byte for byte, as for any document.
- Pages: the document keeps a page map, and each citation names the page of its first and last byte (
Page,EndPage). - Refusals, each a stable code in the ingestion receipt:
pdf-no-text: no page has a text layer, as in a scan; it needs OCR first;pdf-unreadable: the file can't be opened, a page can't be read, or it is encrypted;document-too-large: the file is overMaxFileBytes(64 MiB by default), has more thanMaxPagespages (5,000), or its text is over the ingestion'sMaxDocumentBytes.
A citation proves the quote is in the PDF's extracted text, on the page it names. It doesn't give the quote's position on the page. A PDF ingested without AddGraphRagPdf fails with unsupported-media-type.
Requires Neo4j 2026.09 or later. Guide: PDF. Design: ADR 0052, section 14, in the repository's docs/architecture/adr.
GraphBridge's code is MIT licensed; PdfPig keeps its Apache-2.0 license.
| Product | Versions Compatible and additional computed target framework versions. |
|---|---|
| .NET | net10.0 is compatible. net10.0-android was computed. net10.0-browser was computed. net10.0-ios was computed. net10.0-maccatalyst was computed. net10.0-macos was computed. net10.0-tvos was computed. net10.0-windows was computed. |
-
net10.0
- GraphBridge.GraphRag (>= 0.2.0-preview.1)
- PdfPig (>= 0.1.16)
NuGet packages
This package is not used by any NuGet packages.
GitHub repositories
This package is not used by any popular GitHub repositories.
| Version | Downloads | Last Updated |
|---|---|---|
| 0.2.0-preview.1 | 65 | 10/30/2026 |
Developer preview. Release notes and the breaking-change policy: https://github.com/joslat/GraphBridge.EFCore/blob/main/CHANGELOG.md